Although measurement of both brain structure and function are critically important for linking brain structure and function, these measurements do not appropriately address the important aspect of brain neurotransmission. So far, attempts to map the brain's neurotransmitter systems have largely been limited to histochemical, mRNA-expression or radioligand analyses in rodents and postmortem human brain samples. With the emerging in vivo molecular imaging technologies, in particular positron emission tomography (PET), it has now become possible to image in vivo human brain receptors and transporters from various neurotransmitter systems, at least to the extent that well-validated quantifiable specific PET radiotracers are available. Missing from this approach is the ability to combine PET analysis with cutting edge anatomical and functional analysis to exhaustively characterize receptor binding in terms of its interaction with structure and function. This proposal seeks to combine the analysis tools in FreeSurfer (surfer.nmr.mgh.harvard.edu) created at the Martinos Center in Boston with PET analysis to create atlases of the bind potential of both 5HTT and 5HT4 serotonin receptors. The data to be used comes from the Denmark-based Center for Integrated Molecular Brain Imaging (www.cimbi.org). We will analyze 47 subjects that have both 5HTT and 5HT4 data allowing study of the covariance of these receptors in individuals for the first time. A technique is also proposed to segment the dorsal raphe nucleus from 5-HTT and MRI scans. The functional connectivity of the raphe nucleus will also be studied using this segmentation and resting-state fMRI. We will also create an anatomical probabilistic atlas of the raphe nucleus which can be used to segment raphe in subjects with MRI scans but not PET scans.